Hackpads are smart collaborative documents. .

Vladimir Dubovskiy

1369 days ago
Unfiled. Edited by Vladimir Dubovskiy , Tim Rich , Jim Steinberger 1369 days ago
  • Vlad Dubovskiy, DonorsChoose.org, vlad@donorschoose.org
 
 
 
This is a SAS GeoCoder for use in working with the SAS wrapper...
 
  • The geocoding algorithm lends itself to multi-threading; it could be extended in this way to take advantage of multiple CPUs and run substantially faster
Jim S
  • The current algorithm might be loading the entirety of the relevant CSVs into memory -- if this were recoded to stream and process one line at a time, etc., then the process would no longer require increasing the maximum Java heap size past the default 
 
geocoder_macros.sas - the macros that are were existing from the original. Did not mess with this, but created own file to have a proper home.
 
 
  • Dirty Queens got down and dirty and nailed it. See the parsing model laid out above. (Thanks Shannon, Tommy and Casey!) This is some rock and roll stuff right here!
  • Provided a suggestion roadmap for upkeeping a reference data database for matches to be always compared to a fuller dataset.
 
 
1371 days ago
Unfiled. Edited by Vladimir Dubovskiy 1371 days ago
 
 
  • Skills that you need: Java Programers, any programers really.
  •  
  •  
Vladimir D
  • Where data came from for valid_***.csv inside .JAR file:
  • BBL: PLUTO
  • BIN: PAD
  • X,Y: PLUTO, PAD, CSCL
  •  
  • Parser: Use Melissa data (as it is clean), PAD to do matching for addresses + Reference Data.
 

Contact Support



Please check out our How-to Guide and FAQ first to see if your question is already answered! :)

If you have a feature request, please add it to this pad. Thanks!


Log in